BMW M5 Forum and M6 Forums banner
Cancel
Create thread New Garage Forums

TECH: How to accurately compression test an S62 engine

Jump to Latest
34K views 20 replies 11 participants last post by  Kevin The Clean 1  
#1 · (Edited)
Update Feb 27th, 2007: Added steps 10 a - c, describing how to loosen the sprak plugs and blow clean the combustion chambers to avoid getting soot flakes into the compression tester.

Added 15 b, to remind user to release pressure before disconnecting test hose.

Update Feb 22nd, 2007: Thanks to great input from board member Alts, I´ve corrected important typos in the numbering of the steps in the procedure description, and added some more info.
**** If you have the previous version printed, please replace it with this version ****

How to accurately compression test an S62 engine
======================================

People often get weird results trying to test the cylinder compression pressure on the S62 engine. Here is why it happens, and how to perform the compression test with the best accuracy.

The E39 M5 is specified to have a compression pressure of 12-14 BAR.

The geometric compression is 11.0:1. The compression peak pressures (what is measured during a compression test) is higher because when the air in the cylinder is compressed it gets hot, and this trapped, rapidly compressed air will have a higher temperature and therefor a higher pressure than if it had been compressed slowly and allowed to cool off through heat conduction.

So why the big variation, and why the ultra high pressures often seen in S62B50 compression testing?

The answer is: Fuel.

The fuel system is pressurized even when the fuel pump is not running. The fuel pump is always switched off during compression testing.

The M5 is special. It has a huge fuel pressure, and therefor also a huge residual pressure in the fuel lines when the fuel pump is not running.

The proper way to compression test an M5 is to:

1. Run the engine warm.

2. Stop engine.

3. Pull fuse for fuel pump.

4. Connect a powerful battery charger to keep battery voltage constant.

5. Start engine, and let it stop by itself from fuel starvation.

6. Repeat step 5 until the engine will no longer start.

7. Give full throttle and start engine.

8. Repeat step 7 until the engine does not start.

9. Turn the engine over for a few more seconds.

10 a. Remove all ignition coils. Keep track of what coil goes where.

10 b. Loosen all spark plugs two turns, then retighten.

10 c. Turn the engine over for a few secconds. (this makes any tiny soot flakes broken off from the sparkplug seat fly out the exhaust port, instead of into the compression tester where it can affect meaurements)

10 d. Remove all spark-plugs. Pay attention to keep those in right order.

For instructions how to remove the sparkplugs, see

http://www.m5board.com/vbulletin/showthread.php?t=58768

11. Turn engine over for a few seconds.

12. Repeat until no (or very little) fuel smell comes from the open spark-plug holes. (Don´t breath fuel vapors, it is very bad for you.)

13. Connect compression tester. Firmly tighten the screw-connector into the spark plug hole.
(forget about using a "rubber cone" pressed against the spark-plug hole. Those are a completely joke on a high compression engine...)

14. Press the accellerator pedal completely. Crank engine until you hear EIGHT strokes. (You need this many strokes to get reliable data)

15. Read & note pressure.

15 b. Press pressure release valve on the tester, before you disconnect the hose.

16. Repeat from step 14 for each cylinder in turn.

17. Repeat the whole test one more time from step 14. (You need two sets of cylinder compression data to know you can trust the data.)

18. You now have two sets of compression readings. If there is any significant difference between the readings for any cylinder in the two sets, something is wrong with the test.

It can be a simple thing like a tiny flake of soot coming loose when you removed the sparkplugs.

If the difference is such you want to get more data, then start over from step 11.

19. Remove the battery charger.

20. Put the sparkplugs back in the same locations those came from.

21. Tighten the sparkplugs to factory specification, 30 nm.

For sparkplug instructions see

http://www.m5board.com/vbulletin/showthread.php?t=58768






As an alternative to ventilating all the residual fuel as described above, steps 5-9 can be
replaced by pulling the F9 fuse (feeding the injector valves) located in the engine compartment electronics box, where the DME is, below the right hand passenger compartment intake air filter. IMHO it is easier to vent the residual fuel, than to get access to this fuse.

Looking at posts with compression measurements with very high pressures in some of the cylinders, it is 100% certain the residual fuel was not vented properly before the test commenced, but rater was vented during the test, so the test results are a superposition of the compression characteristics of the engine, the dropping battery voltage (assuming no charger was connected), and the dropping residual fuel pressure.

My advice would be to do the compression test with yourself present and make sure the procedure is followed. or even better spend a few tens of dollars and get the tools needed to do it yourself.

Just make sure to monitor the battery voltage if you don´t have a huge charger connected. If the voltage varies, so will crankshaft speed, and that affects in-cylinder temperature and therefor the measured peak pressures very much. Having a huge charger connected is the best way.

Finally, I would like to re-emphasis the tremendous importance of having the engine fully warm before starting the compression test. With a cold engine the data could be anything, and this would render the data completely irrelevant.

David
 
K
#3 ·
DavidS said:
People often get weird results trying to test the cylinder compression pressure on the S62 engine. Here is why it happens, and how to perform the compression test with the best accuracy.

The answer is: Fuel. David
Click to expand...
David
Once again you have done an excellent job of sharing your knowledge of the M5 with members of the board. Thanks for this outstanding post.

More importantly however you have shown us once again the importance of knowing the ‘mechanics’ and complexity of the M5. Seems that the unknown only frustrates and complicates an owner's (maybe dealers too) ability to properly perform diagnostics and/or make recommendations to correct maintenance problems if you don’t know the whole story.

Ken
 
Save
Reply Quote Like Like
#4 ·
M5_SMG said:
What is the fuel pressure at idle on the S62?
Click to expand...
The fuel pressure is _always_ and _constantly_ regulated to be 5.0 BAR above intake port pressure (the engine vacuum port).

The pressure at the vacuum port varies with engine throttle opening.

At idle the fuel pressure will be 4.5 BAR above ambient. At WOT it will be 5.0 BAR above ambient.

To simulate WOT during fuel pressure testing, simply remove the vacuum hose from the fuel pressure regulator.

David
 
#5 ·
kwalls said:
Seems that the unknown only frustrates and complicates an owner's (maybe dealers too) ability to properly perform diagnostics and/or make recommendations to correct maintenance problems if you don’t know the whole story.

Ken
Click to expand...
Thanks, Ken!

You are definitely right about that, I think, and improving that ability is one of the things this board is all about.

David
 
M
#6 ·
DavidS said:
The fuel pressure is _always_ and _constantly_ regulated to be 5.0 BAR above intake port pressure (the engine vacuum port).

The pressure at the vacuum port varies with engine throttle opening.

At idle the fuel pressure will be 4.5 BAR above ambient. At WOT it will be 5.0 BAR above ambient.

To simulate WOT during fuel pressure testing, simply remove the vacuum hose from the fuel pressure regulator.

David
Click to expand...

Cheers...thanks for that, likes a lot of fuel pressure these cars!
 
K
#8 · (Edited)
1: Run the engine warm.

In case most folks wonder, this is relatively important. The warm temp of the engine vs cold. A warm engine will have higer compressions numbers, generally, than a cold one.

It involves differences in compression due to oil films, and proper running 'sizing/pressuring' of the rings, pistons, and cylinders, from shrinkage/expansion. For example, on my Ford Probe, the last time I did the test, I ran it at idle for 10 minutes to equalize the temperatures in the block and heads. The compression ended up measuring 195 across the board (all 6 cylinders). I had done the test with a cold engine earlier. I measured 160-165 across all 6 cylinders. The numbers were too low, that alarmed me..but I KNEW something was wrong in my test..as they were all nearly the same.

It was lower, and there was variance. In this example, the differences amounted to a nearly 20% variance in compression numbers. Thus, the importance of a warm engine. Not hot, just evenly warm.

I went back and found that I was supposed to kill the fuel pressure, pull the plugs, rotate the warm engine to remove the fuel, untill there was no smell of fuel any more..and then continue with the test. It sounds more involved but really it isn't all that bad, once you are in there.. :)

(*Backyard mechanic warning*)
To add, in case anyone has a Mazda V-6 engine in anything...ALL, I mean ALL of those engines eat oil. From what I hear, the BMW S-62 is designed to eat a bit of oil around the cylinder sleeves...with a film around the rings. This is also true of most engines, is my understanding. The Mazda engine is a bit different. They supposedly have an issue with the softness of the valve stem guides. They will all leak oil around the valve stem seals and guides..and it will pour down the stems into the cylinders, under higher oil pressure, or higher RPM. It makes the oil consumption of the S62 look like a gentle boat ride. I quart per 400km, and even as high as 1 quart per day's 100km blasting, if hammering the high RPM constantly. But..the engines last 300k km (200k miles), no problem. They simply eat oil along the way. One can go through the effort to replace the valve seals/guides, if they wish. This may be nessessary to get the engine to meet emissions testing limits. Just so you know, and don't run off trying to find a new motor for that mazda powered vehicle. The engine has about the same HP/ Displacement ratio as the S-62. 207hp, 2.5 liters displacement. Not bad. Also, a 8000 rpm limit! that is also an extremely repeatable RPM limit, too! As a matter of fact, you are encouraged to beat on this motor, and -mint- used ones are no more than $6-800US, with 40k miles on them!.....:woowoo: What does all this add up to? Beat that Ford Probe/Mazda MX-6 to death! As often as possible! A sure recipe for cheap fun.
 
Save
Reply Quote Like Like
#9 ·
twells said:
David, can you tell me what a "huge battery charger" is? I assume that the cig lighter BMW charger doesn't qualify? Thanks!
Click to expand...
I´d say a 30 amp or better model should do the trick. 30 A will return the battery to normal in half a minute or so after a compression test cranking for one cylinder.

For best cylinder-to-cylinder repeatability watch the voltage of the battery (with the charger switched off) before each cranking.

David
 
#10 ·
KBK said:
1: Run the engine warm.

In case most folks wonder, this is relatively important. The warm temp of the engine vs cold. A warm engine will have higer compressions numbers, generally, than a cold one.

It involves differences in compression due to oil films, and proper running 'sizing/pressuring' of the rings, pistons, and cylinders, from shrinkage/expansion.
Click to expand...
Yes, indeed!

David
 
#14 ·
Updated the text in the first post, please replace any printouts you may have.

David
 
4
#16 ·
David,
One other question. Do you know if VANOS becomes active during cranking while performing a compression test? I know that camshaft profile shifting or valve timing can affect readings obtained during compression test(I'm fairly certain VANOS uses oil pressure for profile change). I have posed this question to one of our FTS(field technical specialist) about CVVT(Volvo's version of variable valve timing) and he could not provide me with a definitive answer. I have experimented with compression tests and have noted a linear drop in compression over all cylinders tested(i.e. start with cylinder 1 to cylinder 5) in an otherwise healthy engine. I thought this might be attributed to a drop in oil pressure during cranking, thus changing valve timing. Any thoughts on this??:dunno:

Thanks again!

Todd
 
#17 ·
Todd,

Accelerator pedal:

I´ve never seen any significant difference between the accelerator pedal pressed or not during compression testing on the S62 engine. I´ve included that particular advice mainly for standardization: most cars need the accelerator pressed during testing, so most mechanics do it always. To make all S62 numbers comparable, I think better everyone do this the same way.

Vanos:

The Vanos units put the camshafts in a good position to start each time the engine is shut off. I assume the cams stay in the same position until the engine runs.

The Vanos units have their own high pressure oil pumps, and also their own pressure accumulator, so the pressure is there immediately once ignition electrical power is on.

Linear drop of compression readings on Volvo engine:

I saw the exact same thing on the S62 engine. I dont think it has anything to do with the oil pressure interacting with the camshaft control.

Then I realized it was fuel being vented into the cylinders, and as the residual pressurized fuel was used up, the amount put into the cylinders lowered, and so did the compression readings. To some extent dropping battery voltage contributed to the drop in the readings.

Those are the very reasons I came up with the procedure described in the first post.

BR,

David
 
4
#18 ·
David,

Agreed on the accelerator pedal. Just good measure even if it does nothing to vehicles equipped with electronic throttle bodies.

Thanks for the information on VANOS and a great deduction on amount of residual fuel in the rail affecting cylinder pressure! That thought has never occured to me. I will have to do a bit more tinkering to see if I can prove the same phenomenon with Volvo.

Thanks again for the excellent information!

Todd
 
#19 ·
Todd,

I think the above procedure (with exception of model specific details, such as spark-plug thread diameter and what number fuses to pull) should work on any car with electronic fuel injection, and improve measurement accuracy.

I believe measurements on an M5 S62 engine with its huge fuel pressure and high-flow fuel injectors benefits comparatively more from this procedure than would measurements on an engine with average fuel pressure.

When you have the results from your Volvo, please post it here.

BR,

David
 
E
#20 ·
I know this post is old but since it's still relevant, I wanted to add some info about the Fuel Pump fuse.

On early model E39 M5s up 2000, specifically up to manufacturing date 09/2000, the Fuel Pump fuse is #54 located on the right side of the trunk behind the trim panel. Once the trim panel is open, the secondary fuse panel is visible. #54 is the second to last on the first set of fuses. I got this info from BMW's WDS or Wiring Diagram System. The is actually a sticker label on the trim panel visible when the trim panel is open with the list of fuses and their function.

I actually pulled fuse #54 after finishing warming up the engine, after 5 seconds or so the engine stopped meaning the fuel pressure was gone at the rail. Tried to start it up a couple of times again until it wouldn't start meaning there wasn't any fuel left at the rail.

Anyway, I just wanted to add this info for early M5s since it's easier to do this than to vent the residual fuel.
 
BMW M5 Forum and M6 Forums
posts
3.2M
members
130K
Since
2000
M5Board is the best forum community for information on the BMW M5 E60 (V-10), E39 (V-8), E34 (straight 6), E28, F90 and F10. Discuss performance, specs, reviews and more!
Full Forum Listing
Explore Our Forums
E39 M5 and E52 Z8 Discussion E60 M5 and E61 M5 Touring Discussion Parts / Other For Sale or Wanted E63/E64 M6 Discussion E34 M5 Discussion